Skip to main content

Full text of "The Latin Prohibitive. Part I"

See other formats


Early Journal Content on JSTOR, Free to Anyone in the World 

This article is one of nearly 500,000 scholarly works digitized and made freely available to everyone in 
the world by JSTOR. 

Known as the Early Journal Content, this set of works include research articles, news, letters, and other 
writings published in more than 200 of the oldest leading academic journals. The works date from the 
mid-seventeenth to the early twentieth centuries. 

We encourage people to read and share the Early Journal Content openly and to tell others that this 
resource exists. People may post this content online or redistribute in any way for non-commercial 

Read more about Early Journal Content at 
journal-content . 

JSTOR is a digital library of academic journals, books, and primary source objects. JSTOR helps people 
discover, use, and build upon a wide range of content through a powerful research and teaching 
platform, and preserves this content for future generations. JSTOR is part of ITHAKA, a not-for-profit 
organization that also includes Ithaka S+R and Portico. For more information about JSTOR, please 


Vol. XV, 2. Whole No. 58. 


Part I. 

This paper owes its origin to a feeling the writer has long had 
that certain uses of the Latin perfect subjunctive are very inade- 
quately and, in some particulars, very inaccurately treated in 
Latin grammars. It is customary, for instance, in dealing with 
ne and the 2d person subjunctive in prohibitions, to dismiss the 
subject with the statement that when the prohibition is addressed 
to no definite person, the present tense is used ; otherwise the 
perfect. All attempts — like Gildersleeve's, 1 for instance — to make 
any further distinction between the tenses have been frowned 
down. Scholars in general have been inclined to accept the 
views of Madvig (Opusc. acad. altera, p. 105) 2 and of Weissen- 
born (on Livy 21, 44, 6) as final, viz. that the perfect is used, 
when a definite person is addressed, only because the present 
cannot be used. The reason for this remarkable state of things 
they do not trouble themselves to seek. Even Schmalz, in the 
second edition of his Lat. Synt., §31, would have it understood 
that the perfect tense in this use has no special significance. 
Such ignoring of all distinction between tenses is common also 
in other constructions, e. g. in the so-called potential subjunctive 

1 Latin Grammar, §266, Rem. 2, which is, as far as it goes, in perfect harmony 
with the results reached in this paper. 

2 Madvig is inexcusably careless in some of his statements in this connec- 
tion. On p. 105, e. g., he says that ne with the present is apud ipsos comicos 
rarissimum et paene inusitatum. As a matter of fact, it is extremely common 
apud comicos — far more so than any other form of prohibition. 


One of the latest grammars (Allen and Greenough, §311) says 
that in aliquis dicat and aliquis dixerit the two tenses refer 
without distinction to the immediate future. The same grammar, 
in dealing with modest assertion, draws no distinction between 
putaverim and putem. It is customary, again, to dismiss the 
perfect subjunctive in prayers with the mere statement that it is 
a reminiscence of archaic formulae, without a hint that the perfect 
necessarily means anything. It has seemed to me that this loose- 
ness of interpretation is entirely at variance with the facts of the 
language, and I have accordingly undertaken an investigation of 
the whole range of those independent constructions of the perfect 
subjunctive in which that tense deals with future time. I have 
included also in my investigation such uses of the future perfect 
indicative as are frequently said to be ' equivalent to the simple 
future.' For the purposes of the paper I have collected and 
classified all the instances of the uses concerned that are to be 
found in all the remains of the Latin language up to the end of 
the Augustan period (except the later inscriptions), together with 
important parts of Silver Latin. I ought perhaps to say that for 
four volumes of the Teubner text I accepted a collection of 
instances made by one of my students. He is, however, one in 
whose care and accuracy I have great confidence, and I feel sure 
that his collection is substantially complete. 

That part of my investigation the results of which I have 
chosen for the present paper deals chiefly with the 2d person, 
present and perfect tenses, of the subjunctive in prohibitions. 
For the purpose of simplifying the discussion I shall, for the 
present, exclude the few cases (commonly called prohibitions 
and classed under ne with the subjunctive) introduced by nee, 
numquatn, nihil (e. g. nee dixeris, nee putaveris). There are so 
serious objections to explaining any one of those introduced by 
nee {neque) in the best prose-writers, and some of those intro- 
duced by nihil, numquam, as instances of the same construction 
as that found in ne feceris, that I shall leave the discussion of 
such cases for Part II of my paper. 

The impression is very generally given that ne with the perfect 
subjunctive is one of the most common methods of expressing 
prohibition in the best classical prose. As a matter of fact, it is 
almost entirely unknown to such prose. It will be understood, 
of course, that the Letters of Cicero do not represent the usage 
of what is understood by 'classical prose.' Tyrrell has clearly 


shown that the diction and constructions in the Letters are the 
diction and constructions of the early comic drama, and not at all 
those of what is commonly meant by Ciceronian Latin. Indeed, 
Cicero himself calls especial attention to the wide difference in 
this respect between them and his other productions in ad fam. 
IX 21, 1 Quid enim simile habet epistola aut iudicio aut contioni? 
. . . Epistolas vero cottidianis verbis texere solemus. We must 
not consider these Letters in determining the usage of the best 
classical prose, any more than we should the usage of early 
comedy : they, as well as the comedy, reflect the language of 
familiar every-day life. Throwing the Letters aside, we may 
say that ne with the 2d person perfect subjunctive does not 
occur in any production, whether prose or poetry, of the whole 
Ciceronian period, except in seven dialogue passages of Cicero 
where the tone distinctly sinks to that of ordinary conversa- 
tion, or unceremonious ordering. 1 If, in addition to these, we 
except four instances in Horace, we may say that it does not 
occur hetween Terence and Livy. It is not to the point to say 
that a prohibition is in its very nature familiar, nor would such 
a statement be true. The orations and the philosophical and 
rhetorical productions of Cicero, as well as the productions of 
other writers belonging to the same period, abound with pro- 
hibitions. The orations of Cicero alone contain 81 prohibitions 
(or probably twice this number if we count such expressions as 
quaeso ne facias, obsecro ne, etc.), and still in his orations no 
instance can be found of ne with the perfect subjunctive except 
in pro Murena 31, where Cicero is quoting the supposed words 
of a teacher to his pupil. 

Again, the grammar-rule which says that the present tense is 
used when the prohibition is general, i. e. addressed to no one in 
particular, while the perfect is used when it is addressed to some 
particular person, or persons, is entirely misleading in the form in 
which it is given. The grain of truth which the rule contains is 
rendered useless by the absence of any hint as to the principle 
involved. Sometimes general prohibitions take the perfect tense, 
e. g. Cato de agri cultura 4 ne siveris ; 37, 1 ne indideris ; 45, 2 
ne feceris; 93 ne addideris; 113, 2 ne siveris; 158, 2 ne addi- 
deris; 161, 2 ne sarueris; XII Tabulae, quoted in Serv. in Verg. 

1 There is no manuscript authority whatever for ne siris (Catullus 66, 91). 
The manuscript reading turn siris is the true one. This matter will be fully 
discussed in Part II of my paper. 


Eel. 8, 99 Unde est in XII tabulis: "Neve alienam segetem 
pellexeris"; Cic. pro Murena 31, 65 Etenim isti ipsi mihi videntur 
vestri praeceptores et virtutis magistri, fines officiorum paulo 
longius, quam natura vellet, protulisse ... " Nihil ignoveris " : 
immo aliquid, non omnia. " Misericordia commotus ne sis": 
etiam, in dissolvenda severitate: sed tamen est laus aliqua 
humanitatis (quoting general precepts of the l vestri praeceptores' • 
which had just been mentioned. Notice the singular verb side 
by side with vestri (instead of tut), which seems to show that the 
prohibition is general) ; Hor. Sat. 2, 2, 16 Quae virtus et quanta, 
boni, sit vivere parvo discite ... hie inpransi mecum disquirite. 
Cur hoc ? Dicam, si potero . . . seu pila velox . . . seu te discus 
agit . . . sperne cibum vilem ; nisi Hymettia mella Falerno ne 
biberis diluta. On the other hand, it is probable that prohibitions 
addressed to definite persons occasionally take the present tense 
at all periods of the literature, and that this use is not, even in 
classical times, confined to poetry, as is commonly supposed. At 
any rate, there are passages in prose which it requires ingenuity 
or violence to explain in any other way, and which, if found in 
Plautus or Terence, no one would have thought of explaining 
in any other way. This use is very common in early comedy, 
and I have collected the following instances from Cicero and later 
prose : Cic. in Verr. II 4, 23, 52 Scuta si quando conquiruntur a 
privatis in bello ac tumultu, tamen homines inviti dant, etsi ad 
salutem communem dari sentiunt. Ne quern putetis sine maximo 
dolore argentum caelatum domo quod alter eriperet protulisse ; 
ib. de republica 6, 12, 12 "St! quaeso," inquit, "ne me e somno 
excitetis et parumper audite cetera" (where the imperative 
'audite' instead of a subordinate subjunctive makes it probable 
that ne excitetis is also independent) ; id. ad fam. 1,9, 23 Quod 
rogas, ut mea tibi scripta mittam, quae post discessum tuom 
scripserim, sunt orationes quaedam, quas Menocrito dabo, neque 
ita multae ; ne pertimescas ; ib. 16, 9, 4 Reliquom est, ut te hoc 
rogem et a te petam : ne temere naviges — solent nautae festinare 
quaestus sui causa — cautus sis, mi Tiro — mare magnum et diffi- 
cile tibi restat — si poteris, cum Mescinio (naviges) — caute is solet 
navigare (where cautus sis and the form taken by the rest of the 
sentence show that ne naviges also is probably independent) ; id. 
ad Att.9, 18,3 "Tu malum," inquies, "actum ne agas" (a proverb 
applied here to a particular person); id. ad Quintum fratrem 1, 
4, 1 Amabo te, mi frater, ne . . . adsignes (Cicero never uses 


amare in this sense with a dependent clause, though its paren- 
thetical use is common in his Letters with independent imperative 
constructions, e. g. ad Att. 2, 2, 1 cura, amabo te, Ciceronem ; 
ib. 16, 16c Amabo te, da mihi et hoc; ib. 10, 10, 3; ad Quint. 2, 
8, [10]) 1 ; Phil. II 5, 10 ne putetis (most naturally taken as inde- 
pendent) ; Livy 44, 22 Vos quae scripsero senatui aut vobis habete 
•pro certis. Rumores credulitate vestra ne alatis, quorum auctor 
nemo exstabit (This, or some reading which involves the same 
construction, seems inevitably correct, and would undoubtedly 
be accepted by everybody were it not for the supposed rule) ; 
ib. 22, 39, 2 Armatus intentusque sis, neque occasioni tuae desis 
neque suam occasionem hosti des (Livy and later writers freely 
use neque for neve); Tac. Dialogus 17 Ex quo colligi potest et 
Corvinum ab illis et Asinium audiri potuisse (nam Corvinus in 
medium usque Augusti principatum, Asinius paene ad extremum 
duravit). Ne dividatis saeculum, et antiquos ac veteres vocitetis 
oratores quos eorundem hominum aures adgnoscere ac velut 
coniungere et copulare potuerunt. It was formerly customary 
among editors of the Dialogus to punctuate this sentence as 
above. Recent editors use only a comma or a semicolon before 
ne dividatis, understand an ellipsis (i. e. Haec dico ne, etc.), and 
thus make Tacitus use a very awkward sentence. Why make 
this so difficult ? Why not let it be what it seems to be on the 
face of it, namely, a prohibition? 

Here, then, are several instances in prose of the present subjunc- 
tive with ne addressed to a definite person. The reason why it 
is not more common will appear later on in this discussion. But 
even if none of these examples existed (and there have been 
ingenious attempts to explain away most of them in deference 
to the supposed rule), there would still be no ground for such a 
rule. In the whole field of classical prose from the beginning of 
the Ciceronian period to the end of the Augustan period, and 
even later, there is but a single example of ne with the indefinite 
2d person present subjunctive in a prohibition. There are a few 
examples from poetry, but these have no bearing upon the point 
in question, as it is everywhere acknowledged that ne with the 
present is common in poetry even in addressing a definite person. 
The single example just referred to is of course the one cited 
under this rule, with suspicious uniformity, by all Latin gram- 

1 Even in Plautus and Terence amabo in this sense is almost invariably 
thrown in parenthetically. 


mars, viz. Cic. Cato Maior 10, 33, though even here it might be 
noticed that Cato is speaking to definite persons, addressing at 
one time Scipio individually, again Laelius, and still again both 
together. The truth is that a general prohibition in Latin is 
nearly always expressed by the use of the 3d person, e. g. ne 
quis putet, etc., or some circumlocution introduced by cavendum 
est ne, or the like. It will, I think, be admitted that the above 
considerations at least cast serious doubt upon the validity 
of the grammar-rules regarding the use of ne in prohibitions. 
The question as to the true distinction between the tenses in such 
constructions seems to me to be still an open one, and this paper 
is intended as a contribution to its solution. 

Let us start with certain general principles. All will agree 
that the perfect subjunctive, when dealing with a future act, 
differs, at least in some uses, from the present in representing 
the act as one finished in the future. For instance, in the expres- 
sion si venerit, videat the act of coming is conceived of as a 
finished act in the future, about to be completed prior to the 
beginning of the act of seeing. In si veniat, on the other hand, 
the act is conceived of as in progress in the future. Such a 
distinction between the tenses of nefeceris and ne facias would 
not be entirely satisfactory at all points of the parallel. Ne 
feceris cannot mean literally ' Do not prior to a certain point in 
the future, have done it.' In one respect, however, the distinc- 
tion, it seems to me, still holds. In ne feceris there is at least no 
thought of the progress of the act. The expression deals with 
an act in its entirety. The beginning, the progress and the end 
of the act are brought together and focussed in a single concep- 
tion. The idea of the act is not dwelt upon, but merely touched, 
for an instant, and then dismissed. The speaker, as it were, 
makes short work of the thought. There is a certain impetus 
about the tense. When a man says ne facias he is taking a 
comparatively calm, dispassionate view of an act conceived of 
as one that will possibly be taking place in the future ; nefeceris, 
on the other hand, implies that the speaker cannot abide the 
thought ; he refers to it only for the purpose of insisting that it 
be dismissed absolutely as one not to be harbored. As far as 
the comparative vigor of the two expressions is concerned, the 
difference in feeling between them is similar to that between 
' Go ! ' and ' Be gone ! ' ' Go ' dwells upon the progress of the 
act. A man never says 'Be gone!' except when aroused by 


strong emotion, which does not allow him to think of the 
progress of the act, but only the prompt accomplishment of it 
In a similar way nefeceris betrays stronger feeling than ne facias 
— it disposes of the thought with the least possible ado. The 
same distinction should be made between cave feceris and cave 
facias. This feature of the tense, if my characterization of it is 
correct, would lead us to expect it to be used only, or chiefly, in 
animated, emotional, or unusually earnest discourse, and to such 
passages, as we shall presently see, is it almost exclusively con- 
fined. I wish to insist upon this as the only real distinction 
between the two tenses with ne. We shall now, of course, expect 
that in the majority of cases where a prohibition is a general, 
indefinite one, the present tense will be found. When a man is 
soberly philosophizing and writing precepts for the world at 
large, he is not often aroused by emotions so strong as he is 
when, actually face to face with a person and perhaps under the 
influence of anger, alarm or some other intense feeling, he orders 
that person not to do a certain thing. But even in this sort of 
writing, when he feels that his precept is of prime importance, he 
may occasionally fall into the more vigorous form of expression. 
For the satisfactory study of such expressions we look for some 
production abounding in general precepts, and still not written in 
the form of dialogue and not addressed to any one in particular. 
Naturally we turn to Cato's de agri cultura. In the seven 
different passages of this work cited above, Cato uses ne with 
the perfect in a general prohibition. In each case the context 
makes it probable, or, in the light of facts which I shall present 
later, practically certain, that he considers of especial importance 
the particular thing prohibited, e. g. ch. 4, where he is trying to 
show how a farmer may live happy and prosperous : ruri si recte 
habitaveris, libentius venies : fundus melior, minus peccabitur, 
fructi plus capies. Frons occipitio prior est : vicinis bonus esto : 
familiam ne siveris peccare. Si te libenter vicinitas videbit, facilius 
tua vendes, operas facilius locabis etc., i.e. 'above all things, do not 
allow the members of your household to offend them. If you keep 
on good terms with your neighbors, you will find it easier to sell 
your produce,' etc.; again, 37, 1 : 'If you are dealing with land that 
is cariosa, peas are a bad crop to put in ; so are barley, hay, etc.; 
above all things, do not put in nuts (nucleos ne indideris).' Every- 
where else in his treatise he uses the less vigorous forms of prohi- 
bition, sometimes nolito with the infinitive, sometimes ne with the 


2d imperative, sometimes caveto with the present tense of the sub- 
junctive. He never uses the perfect tense with caveto, though this 
tense with cave is far more common in Plautus than the present. 
The present tense, on the other hand, occurs in Cato 17 times. 

By far the best place to study the difference in meaning between 
the two tenses is in Plautus and Terence, because in them (and 
only in them) both tenses are very freely used with ne and cave 
in prohibitions. It is there, too, that the tone of the prohibition 
can best be determined, because the dramatic action makes clear 
the feeling of the speaker. I give below classified lists of all the 
passages in Plautus and Terence containing prohibitions of this 
sort. 1 In studying these lists, there are certain considerations 
which should be kept constantly in mind. In all but a compara- 
tively few cases, the distinction I have drawn between the perfect 
and the present tenses will be very clear. But of course some 
instances, both of the perfect and of the present, will be found 
near the border-line. In some cases where the speaker is moved 
by only slight emotion, one tense would be as appropriate and 
natural as the other. Again, a speaker may be somewhat aroused 
while still under perfect self-control and realizing the advisability 
of calm language. On the other hand, a speaker may be really 
very calm, while wishing, for certain purposes, to seem very 
indignant. We should also bear in mind a natural tendency to 
unceremoniousness and a vigorous off-hand style in every-day 
conversation between friends and in the language of superiors to 
inferiors. If we keep in mind these considerations, a comparison 
of the following lists will, I think, inevitably lead to the conclusion 
that the distinction I have drawn is the true one. 

There are in Plautus and Terence 31 instances of ne with the 
perfect subjunctive. In nearly all of these the feeling of strong 
emotion of some sort — e. g. great alarm, fear of disaster if the 
prohibition is not complied with, or the like — is very prominent. 
Many of them are accompanied by other expressions which 
betray the speaker's earnestness, e. g. per deos atque homines, 
opsecro, hercle, etc. And there is not one of them in the least 
inconsistent with my explanation of the meaning of the tense. 
Plautus has this construction in the following passages 2 : Am. 924 

1 1 was surprised to find no instance of this use in the tragedies of Seneca, 
who, I believe, uses only ne with the imperative (or vide ne with the subjunc- 
tive) in prohibitions. 

2 1 have not thought it necessary for my present purpose to make a separate 
class of such aorists as dixis, parsis, etc. 


Per dexteram tuam te, Alcumena, oro, opsecro te, da mi hanc 
veniam, irata ne sies (evidendy here the perfect of irascor. The 
fact that this verb is inchoative in form does not militate against 
the principle I have laid down, as it is seldom inchoative [never 
so, if we may trust Harpers' Diet.] in meaning. It commonly 
means to feel angry. When the beginning of the act is referred 
to incipio, or a verb of similar meaning is used with it, e. g. ad Att. 
4, 1, 8 incipiunt irasci. Inchoative verbs are not found in this 
construction) ; Miles 283 Sc. Nescis tu fortasse, apud nos facinus 
quod natumst novom. Pal. Quod id est facinus? Sc. Impudicum. 
Pal. (not wanting to hear such news) Tute sci soli tibi : Mihi ne 
dixis. Notice the many indications of earnest feeling: Tute (tu 
alone even would have been emphatic) soli tibi, and all sharply 
contrasted with mihi; ib. 862 Perii: excruciabit me erus . . . Fu- 
giam hercle . . . ne dixerilis, opsecro, huic vostram fidem ! ib. 
r 333 : Here Philocomasium has just fainted and fallen into the arms 
of her lover, at the thought of leaving him. All is excitement. 
One says : Run for some water. The lover exclaims : ne inter- 
veners, quaeso, dum resipiscit; Rudens 11 55 Perii in primo 
praelio: mane! ne ostenderis! Here his possession of the 
treasure that has been found depends, as he thinks, upon its 
not being shown ; Trin. 521 Per deos atque homines dico, ne 
tu illunc agrum tuom siris umquam fieri ; ib. 704 (Lysiteles in a 
quarrel with Lesbonicus, indignant at the suggestion of anything 
which might reflect upon his character) Id me commissurum ut 
patiar fieri ne animum induxeris ; ib. 1012 Ne destiteris currere 
(addressed to himself in fear of a flogging. All his words at this 
point indicate hurry and alarm); Asin. 839 Son (in a tone of 
earnest deprecation, in answer to his father's taunt) : Ne dixis 
istuc. Father: Ne sic fueris: ilico ego non dixero; Cure. 599 
Planesium (to Phaedromus, in great fear lest the parasite escape 
with the stolen ring) . . . propera! . . . Parasitum ne amiseris! 
Pseud. 79 Id quidem hercle ne par sis! Most. 1083 Theopro- 
PIDES (angry, and resolved to punish Tranio, trying to get him 
away from the altar, where he had taken refuge) : Surge . . . ne 
occupassis, opsecro, aram . . . surgedum hinc . . . surge : ne 
nugare. Aspicedum; Men. 415 Ne feceris! periisti, si intrassis 
intra limen ; ib. 617 Pe. (during an angry dispute) At tu ne clam 
me commessis prandium. Me. Non taces? Pe. Non hercle 
vero taceo; Epid. 150 (in answer to Stratippocles' intimation 
that he would commit suicide) ne feceris! ib. 593 Per. Si hercle 


te umquam audivero me patrem vocare, vitam tuam ego interi- 
mam. Fid. Nonvoco . . . nefueris pater; Poen. 552 (the lawyers, 
speaking with professional decisiveness and importance) Nos tu 
ne curassis! scimus rem omnem. The tone assumed here by 
the speakers may be inferred from the fact that they have just 
been accused of speaking with too much anger (cf. vs. 540 nimis 
iracundi estis) ; ib. 990 ne parseris; Aul. 100 (Euclio having a 
large amount of gold concealed in his house, is constantly alarmed 
lest it be stolen. He bids his servant again and again not, under 
any circumstances, to let any one enter the house) Si bona 
Fortuna veniat, ne iniromiseris ! ib. 577 Euc. (still in fear of 
losing his treasure) Ne in me mutassis nomen ! ib. 737 Lyc. 
(upon Euclio's threatening him with death) Ne istuc dixis / ib. 
790 Ne me uno digito adtigeris, ne te ad terram, scelus, adfligam ! 
Cas. 2, 6, 52 St. Praecide os tu illi! Age! Cle. (trying to 
prevent a fight) Ne obiexis manum ! Cist. 1, 1, 111 Silenium 
(speaking of her lover, with great depth of feeling that moves 
her hearers to tears [vs. 113]) sed, amabo, tranquille; ne quid, 
quod illi doleat, dixeris ! The following seems near the border- 
line, one tense being as appropriate as the other : Merc. 396 ne 
duas neu te advexisse dixeris. 

Terence has only two instances of ne with the perfect : Phorm. 
514 Unam praeterea horam ne oppertus sies. The speaker is 
fairly beside himself throughout this scene, which sufficiently 
accounts for the more emotional form of expression. Ib. 742 
(alarmed by fear lest his treachery be discovered) Ne me istoc 
posthac nomine appellassis. 

The same feeling that prompts the use of the perfect tense in 
the passages just cited, explains the use of the same tense in 
prohibitions introduced by cave. Plautus and Terence present 
33 instances of cave with the perfect: Plaut. Am. 608; Miles 
1 125; 1245; 1368; 1372; Trin. 513; 555; Asin. 256; 467; 625; 
Bacch. 402; 910; 1 188; Stich. 284; Most. 388 ; 508; 795; Men. 
996; Epid. 400; 434; Merc. 112; 476; Poen. 1020; Aul. 90; 
600; 610; Persa 388; 933; Cas. II 5, 24; Ter. And. 753; 760; 
Haut. 187 ; Adelph. 458. 

If now we turn to ne and cave with the present subjunctive we 
find a very different state of things. There are in Plautus and 
Terence more than 100 instances of ne, and 18 (19?) instances of 
cave, in this form of prohibition, as will be seen by consulting the 
following list : Am. 87 (Prologue addressing the audience) Mirari 


nolim vos, quapropter Juppiter nunc histriones curet. Ne mire- 
mini 1 : ipse hanc acturust Juppiter comoediam; ib. 116 (still 
addressing the audience) Ne hunc ornatum meum admiremini ; 
Capt. 14 Ego me tua causa, ne erres, non rupturus sum (probably 
ne here means 'lest'); ib. 58 (Prologue) Ne vereamini, quia 
bellum Aetolis esse dixi cum Aleis; ib. 186: The parasite 
(replying to Hegio, who has good-humoredly warned him not 
to expect too much at his table) : Numquam istoc vinces me, 
Hegio: ne postules cum calceatis dentibus veniam ; ib. 331 Filius 
meus aput vos servit captus : eum si reddis mihi, praeterea unum 
nummum ne duis; ib. 349 Nee quemquam potes mittere ad eum 
quoi tuom concredat filium audacius. Ne vereare: meo periculo 
ego huius experiar fidem ; ib. 393 Istuc ne praecipias, facile 
memoria memini ; ib. 854 Nee nihil hodie nee multo plus tu hie 
edes, nefrustra sis; ib. 947 At ob earn rem mihi libellam pro eo 
argenti ne duis: gratiis a me ducito; ib. 957 Fui . . . bonus vir 
numquam neque frugi bonae neque ero umquam : ne spem ponas 
me bonae frugi fore; Miles 1215 Py. Libertatem tibi ego et divi- 
tias dabo, si impetras. Pa. Reddam impetratam ... At modiee 
decet. Ne sis cupidus; ib. 1274 Viri quoque armati idem istuc 
faciunt: ne tu mirere mulierem; ib. 1360 Pa. Muliebres mores 
discendi. Py. Fac sis frugi. Pa. lam non possum: amisi 
omnem lubidinem. Py. I, sequere illos: ne morere; ib. 1378 Ne 
me moneatis: memini ego officium meum; ib. 1422 Aliter hinc 
non ibis: ne sis frustra ; Rud.941 Nil habeo, adulescens, piscium : 
netu mihi esse postules ; ib. 968 Gr. Hunc homo nemo a me feret ; 
ne tu te speres. Tr. Non ferat, si dominus veniat? Gr. Domi- 
nus huic, ne (probably = 'lest') frustra sis, nisi ego nemo natust, 
hunc qui cepi in venatu meo ; ib. 992 Quod in mari non natumst 
neque habet squamas ne /eras; ib. 1012 Hinc tu nisi malum 
frunisci nil potes, ne postules ; ib. 1368 Ut scias gaudere me, mihi 
triobulum ob earn ne duis; ib. 1385 Quod servo meo promisisti, 
meum esse oportet. Ne tu, leno, postules ; ib. 1390 Dae. Opera 
mea haec tibi sunt servata: (Gr. Immo hercle mea, ne tu tua 
dicas) ; ib. 1414 nihil hercle hie tibi, ne tu speres; Trin. 16 
(Prologue, to audience) de argumento ne expectetis fabulae ; ib. 
267 Apage sis amor. Amor, amicus mihi ne fuas umquam ; ib. 
370 Ph. . . . quid dare illi nunc vis? Lu. Nil quicquam, pater: 
Tu modo ne me prohibeas accipere, siquid det mihi ; Bacch. 747 

'Some of these might be explained as final clauses ('that you may not be 
surprised,' I make the following statement, etc.). 


. . . quod promisisti mihi te quaeso ut memineris, ne ilium verberes 
(probably a dependent clause) ; ib. 758 . . . ubi erit adcubitum 
semel, ne quoquam exurgatis, donee a me erit signum datum ; 
Cure. 183 Pa. Quin tu is dormitum? Ph. Dormio: ne occla- 
mites; ib. 213 Si amas, erne: ne rogites; ib. 539 Ne mihi te 
facias ferocem aut supplicare censeas; ib. 565 Nil aput me 
quidem. Ne facias testis : neque equidem dehibeo quicquam ; 
ib. 568 Vapulare ego te vehementer iubeo: ne me territes ; ib. 
713 Non ego te flocci facio; ne me territes (the feeling in such 
cases is not that the failure to comply with 'ne territes' will be 
disastrous to me, but that it will do you no good to try to frighten 
me); Ps. 275 . . . scimus nos te qualis sis: ne praedices ; ib. 1234 
Sequere tu. Nunc ne expectetis, dum domum redeam ; Stich. 
320 Tua quod nil refert, ne cures ; ib. 446 ... id ne vos mire- 
mini, homines servolos potare etc.; Most. 598 Pater advenit . . . : 
is tibi et faenus et sortem dabit. Ne inconciliare nos porro 
postules; ib. 611 Tra. Huic debet Philolaches paulum. Theop. 
Quantillum ? Tra. Quadraginta minas. Theop. Paulum id 
quidemst? Tra. Ne sane id multum censeas; ib. 799 Ergo 
inridere ne videare et gestire admodum ; ib. 994 Ad cenam ne 
me te vocare censeas; ib. 1010 Theop. Minas tibi octoginta 
argenti debeo. Si. Non mihi quidem hercle: verum si debes, 
cedo. . . . Ne ire initias postules ; Men. 327 ne quo abeas longius 
ab aedibus; ib. 790 Quid ille faciat, ne id observes; Epid. 147 
Ep. A quo trapezita peto ? Strat. Unde lubet. Nam ni . . . 
(prompseris), meam domum ne inbitas; ib. 305 Ne abitas, prius- 
quam ego ad te venero ; ib. 339 [hoc quidem iam periit, ne quid 
tibi hinc in spem referas (perhaps dependent)] ; Merc. 164 
Char. Quid istuc est mali? Acan. Ne rogites; ib. 318 Dem. 
Ne me obiurga. Lys. . . . non obiurgo. Dem. At ne deteriorem 
hoc facto ducas (there seems to be slight emotion here; either 
tense would seem appropriate) ; ib. 396 Ne duas neu te advexisse 
dixeris (this, like the passage just cited (vs. 318), seems on the 
border-line. The speaker is really very earnest, but is, as shown 
by the general situation, anxious not to appear too much so, lest 
his real motive be guessed. The sudden change of tense, then, 
is not surprising) ; ib. 457 Ad portum ne bitas, dico iam tibi 
(perhaps dependent) ; ib. 520 Nunc, mulier, ne tu frustra sis, 
mea non es ; ne arbitrere; Poen. 520 Ne tuo nos amori servos 
esse addictos censeas; ib. 526 Afetu opinere (perhaps dependent); 
ib. 536 Est domi, quod edimus, ne nos tam contemptim conteras 


(perhaps dependent upon 'I say this,' understood); ib. 1152 
Audin tu, patrue ? Dico, ne dictum neges (perhaps dependent) ; 
ib. 1370 Ne mirere, mulieres, quod eum sequontur; Aul. 166 
Verba ne facias, soror; ib. 231 Eucl. At nihil est dotis quod 
dem. Meg. Ne duas, dum modo morata recte veniat, dotatast 
satis. Eucl. Eo dico, ne me thensauros repperisse censeas. 
Meg. Novi; ne doceas; ib. 350 Sunt igitur ligna, ne quaeras 
foris ; Persa 141 Numquam hercle hodie hie prius edis, ne 
frustra sis; True. 477 Ne exspectetis, spectatores, meas pugnas 
dum praedicem ; ib. 658 Ne me morari censeas ; ib. 744 Res ita 
est, ne frustra sis; Cas. Prol. 64 (to audience) Ne exspectetis 
etc.; ib. II 6, 42 Ne a me memores malitiose de hac re factum, 
aut suspices; Cist. II 3, 16 Nam illaec tibi nutrix est: ne matrem 
censeas; ib. V (to audience) Ne expectetis, spectatores etc. In 
Capt. 548 Hegio, hie homo rabiosus habitus est in Alide : ne tu 
quod istic fabuletur auris inmittas tuas, and in Miles 1363 (1351) 
Pa. Si forte liber fieri occeperim mittam nuntium ad te : ne me 
deseras, there seems to be a certain amount of emotion, but it 
will be noticed that in each case the speaker is addressing a 
superior. In the former case, too, the speaker is anxious to 
appear calm and undisturbed. Furthermore, ne might well be 
taken here in the sense of 'lest.' In the other passage, the slave 
who is speaking does not even mean what he says. He is really 
glad that he is going, and never wants to see again the master 
whom he is addressing. In the light of this fact, ne deseras 
seems cool irony. The stereotyped formula ne molestus sis 
occurs in Plaut. Asin. 469; Ps. 118; 889; Most. 74; 572; 757; 
863; 871; Men. 251; Aul. 450; but in nearly all of these 
instances it might be taken as dependent upon some other verb 
expressed or understood. In any case, one must not look for 
strong emotion in so commonplace a phrase. Ne with the 
present subjunctive occurs in Terence in the following passages : 
And. 704 Huic, non tibi, habeo, ne erres (perhaps dependent) ; 
ib. 706 Dies hie mihi ut satis sit vereor ad agendum : ne vacuom 
esse me nunc ad narrandum credas; ib. 980 (to audience) Ne 
exspectetis dum exeant hue; Eun. 76 Quid agas? nisi ut te 
redimas captum quam queas minimo : . . . et ne te adftictes ; 
ib. 212 Ego quoque una pereo, quod mihi est carius: ne istuc 
tarn iniquo paiiare animo; ib. 273 Gn. Quia tristis es. Pa. 
Nihil quidem. Gn. Ne sis; ib. 388 Si certumst facere, faciam : 
verum ne post conferas culpam in me; ib. 786 Sane quod tibi 


nunc vir videatur esse hie, nebulo magnus est: ne meluas ; 
ib. 988 Ere, ne me spectes: me inpulsore haec non facit; Haut. 
745 Sy. Ancillas . . . traduce hue propere. Dr. Quam ob rem ? 
Sy. Ne quaeras; Phorm. 419 "Actum" aiunt "ne agas"; Hec. 
342 Non visas ? Ne mittas quidem visendi causa quemquam ; 
Adelph. 22 Ne exspectetis argumentum fabulae. In Phorm. 508 
Heia, ne parum leno sies, the «^-clause is rightly explained by 
editors as dependent ' Look out there, lest,' etc. Besides these, 
there are five instances of ne attigas which will call for comment 

Cave with the present tense of the subjunctive occurs as follows : 
Plaut. Capt. 431; 439; Most. 797; 1012; Epid. 432; Persa 52; 
812; Cas. Ill 1, 16; Poen. 117; Ter. Eun. 751; Haut. 302; 
826 (?) ; Phorm. 993; Adelph. 170. 

There are certain remarkable differences between the prohi- 
bitions in this latter list (expressed by the present tense) and 
those in the former list (expressed by the perfect) which a casual 
observer might not notice. If my distinction between the two 
tenses is correct, we should expect that a prohibition dealing with 
mere mental action, e. g. 'Do not suppose,' 'Do not be surprised,' 
'Do not be afraid,' would commonly take the present tense, 
because such prohibitions would not commonly be accompanied 
by strong emotion, and, as far as the interests of the speaker are 
concerned, it matters little whether the prohibition be complied 
with or not. Such a condition of things is exactly what we find. 
Among the instances of ne with the perfect tense, not a single 
example of a verb of this class will be found; but among those 
of ne with the present there are no less than 31 instances of such 
verbs, or nearly a third of the entire number. Again, such 
prohibitions as 'Do not ask me,' 'Do not remind me' (i.e. 
I know already), would not ordinarily imply any emotion, and 
no such verbs will be found among the instances of ne with the 
perfect. 1 But there are 13 such verbs among the instances of the 
present. Substantially the same holds true for the taw-construc- 
tions. Among the 33 instances of cave with the perfect there is 
no instance of a verb belonging to any of these classes. There 
is no avoidance of such verbs with cave used with the present 

1 The nearest approach to an exception is iratus ne sies (Plaut. Am. 924), 
which seems here to be the perfect tense of irascor. Here there is an addi- 
tional idea of venting one's anger, which removes it, strictly speaking, from 
the class referred to. 


tense (in spite of the fact that there are only about half so 
many instances of the present as of the perfect), e. g. Ter. 
Phorm. 993 ; Haut. 826 (admiralus here probably used adjec- 
tively, as in ad Att. 9, 12, 2 and Off. 2, 10, 35) ; Plaut. Asin. 
372; Capt. 431 (?) ; or with noli (though noli is comparatively 
rare in Plautus and Terence), e. g. Plaut. Persa 619; Capt. 
845 ; Ter. Phorm. 556 ; or with ne followed by the imperative, a 
construction which occurs 33 times in Plautus and Terence with 
such verbs (out of a total of 84 instances) : Plaut. Am. 674 ; 
1064; 1 1 10; Capt. 554; Miles 893; 895; ion; 1345; Rud. 688; 
1049; Trin. 1181; Asin. 462; 638; 826; Cure. 520; Ps. 103; 
734; 922; Men. 140; Merc 172; 873; 879; 993; Cas. 4, 4, 14; 
Most. 629 ; True. 496 ; Aul. 427 ; Persa 674 ; Ter. And. 543 ; 
Adelph. 279 ; 942 ; Haut. 85 {bis)} Outside of Plautus and 
Terence such verbs occur, in the ante-Ciceronian period, as 
follows: Cato de agr. cult. 1, 4 caveto contemnas; ib. 64, 1 nolito 
credere (' do not believe ') ; Corpus Inscriptionum Latinarum, 1 1445 
credere noli; ib. 1453 spernere nolei. But nowhere in this whole 
period is such a verb to be found in the perfect tense in a prohi- 
bition. Why this mysterious absence of all such verbs from this 
one sort of prohibition? Recurring to the instances of the 
present tense in Plautus and Terence, we notice that in n of 
the passages the prologue, or some one else, is calmly addressing 
the audience with ' Do not expect me to disclose the plot of the 
play,' or some prohibition equally calm. But there is not one 
instance in the prologues either of Plautus or Terence of the 

1 It will be noticed that in Plautus and Terence more than one-third of the 
verbs in prohibitions expressed by ne and the imperative are verbs of fearing 
(22 of the 33), thinking, asking or advising. Of the remaining verbs, a large 
proportion are verbs of saying and weeping. A similar state of things prevails 
in Vergil, who uses this construction 27 times. In 12 of these the verbs 
belong to the classes just mentioned. All this is interesting in connection 
with the much-mooted question regarding the relative harshness in Greek of 
fir/ with the present imperative and jit) with the aorist subjunctive. See Dr. 
Miller's paper on the Imperative in the Attic Orators, A. J. P. XIII 424. In 
Latin, ne with the perfect subjunctive is harsher than ne with the imperative, 
the latter corresponding rather closely in this respect with ne and the present 
subjunctive. Both of these last-mentioned constructions, however (ne with 
imperat. and ne with pres. subj.), smacked somewhat of the same familiar 
feeling as their sister construction. Noli was far more deferential, and 
Cicero, when he wished to soften the tone of his address, accordingly 
preferred that form of prohibition. 


perfect tense in prohibition. And this again is exactly what we 
should expect. (It matters little for our present purpose whether 
Plautus wrote the prologues to his plays or not.) In general the 
fact may be emphasized that ne with the present is chiefly 
confined to prohibitions of the most commonplace sort. Where 
this is not apparent from the nature of the verb itself, a study of 
the context will show that the speaker is not under the influence 
of any strong emotion. There are in all only 5 instances (a small 
number out of so many) which can fairly be said to be accom- 
panied by decided emotion, and in each case, strangely enough, 
the verb is attigas, viz. Plaut. Bacch. 445 ; Most. 453 ; Epid. 721 ; 
True. 273; Ten And. 789. I cannot account for this strange 
exception, unless one accepts Curtius' suggestion that attigas is 
an aoristic form (Stud. V 433). The few additional passages 
that might apparently be construed as exceptions have been 
commented upon under the citation. 

Whatever differences of opinion may be held regarding indi- 
vidual instances in the two lists above given, I feel sure that no 
one who studies them carefully can resist the general conclusion 
to which I have come. If, now, the distinction I have drawn 
between the two tenses holds so clearly for the only two authors 
who make frequent use of ne with the subjunctive in prohibitions, 
a strong presumption is established in favor of a similar distinc- 
tion in the few instances to be found in later writers, where there 
are not always so many indications at hand, as in dramatic pro- 
ductions, to make clear the feeling of the writer. And a study 
of these instances confirms the presumption. There are in 
classical prose, from the beginning of the Ciceronian period up 
to near the end of the Augustan period, only seven instances of 
ne with the perfect in prohibition, and these are all in Cicero. 
As pointed out above, each of these occurs in dialogue where 
the tone sinks to that of ordinary conversation, in which some 
one is delivering himself of an earnest, energetic command. One 
is naturally more unceremonious in addressing a familiar friend 
than in addressing a mere acquaintance: he falls more readily 
into energetic forms of expression. Often he assumes an off- 
hand, imperious tone in such cases merely as a bit of pleasantry. 
This would be especially natural when one was urging his friend- 
not to do what he feared that friend might do — namely, in pro- 
hibitions. One can hardly fail to notice this tone at any talkative 
gathering of intimate friends. Let us examine now more care- 


fully the seven instances referred to: de div. 2,61, 127 (a sup- 
posed command of a god to a man) hoc nefeceris! de rep. 1, 19, 
32 Si me audietis, adulescentes, solem alterum ne metueriiis! de 
leg. 2, 15, 36 (Atticus, replying sharply to Marcus) Tu vero istam 
Romae legem rogato: nobis nostras ne ademeris! Ac. 2, 40, 125 
(in conversation with Lucullus at a familiar gathering of friends) 
Tu vero ista ne asciveris neve fueris commenticiis rebus adsensus ! 
Tusc. disp. 1, 47, 112 (replying in a deprecatory tone to a sug- 
gestion that has just been made) Tu vero istam ne reliqueris! 
pro Mur. 31, 65 (quotation from the supposed command of a 
teacher to his pupil) misericordia commotus ne sis! (though sis 
alone might be looked upon as the verb here, in which case the 
construction would belong to the other class); Par. Sto. 5, 3, 41 
(in a vigorous protest) tu posse te dicito, debere ne dixeris. An 
unusually earnest and energetic tone is to be found in each one 
of these. Notice, for instance, the strongly contrasted pronouns 
and the other indications of strong feeling. The reason why this 
construction is so rare in classical productions is that they are, for 
the most part, of a very dignified character. The prohibitions 
they contain are therefore commonly expressed by noli with the 
infinitive (a construction that occurs 123 times in Cicero, twice in 
Nepos, three times in Sallust, three times in Caesar), or by cave 
with the present subjunctive (30 times in Cicero, once in Nepos, 
once in Sallust), or by vide ne with the subjunctive (18 times in 
Cicero, once in Nepos). Next to noli, the most common form of 
prohibition in Cicero is, I should say, some circumlocution like 
pelo, rogo, oro, etc., followed by ne and the subjunctive, but I 
have made no attempt to collect the instances. Even ne with 
the present subjunctive is less deferential than the constructions 
just named ; it smacks somewhat of its sister construction, and so 
is comparatively rare. Where, next to the early comedy, do we 
find the most familiar tone prevailing ? One may answer, without 
hesitation, in the Letters of Cicero. And it is in these Letters 
that most of the instances of ne with the perfect in classical times 
are found. It is also a significant fact, and one, I think, not 
hitherto noticed, that all but 2 of the 14 instances here found 
are addressed to his bosom-friends or relatives: 8 of them to 
Atticus, 2 to his brother Quintus, and 2 to his intimate legal 
friend Trebatius, upon whom he was always sharpening his wits 
and whom he never lost an opportunity to abuse, good-naturedly, 
to his face. One of the -two exceptions is in a very impassioned 


passage of a letter written by Brutus to Cicero, ad Brut, i, 16, 6 ; 
the other is in ad fam. 7, 25, 2, where Cicero is enjoining upon 
Fadius Gallus, in the most urgent terms possible, not under any 
circumstances to reveal a certain secret. To his other corres- 
pondents he uses only noli or, in two instances, cave with the 
present subjunctive, e. g. to Servius Sulpicius (ad fam. 4, 4, 3), to 
Lucius Mescinius (ad fam. 5, 21, 1), to Cornificius (ad fam. 12, 30, 
1 ; 12, 30, 3), to Gallus (ad fam. 7, 25, 1 ; 7, 25, 2), to Brut, i, 6 
twice; 1, 7 ; 1, 13; 1, 15, 1 twice, etc. Excepting the passionate 
remonstrance referred to in a letter written by Brutus, the corres- 
pondents of Cicero use only noli when addressing him, e. g. ad 
fam. 4, 5, 5; 7, 29; 12, 16, 1. In the treatise ad Herennium, I 
might add, ne never occurs in prohibition, though other forms of 
prohibition are common, e. g. noli in 4, 30, 41 ; 4, 41, 53 twice; 
4, 52, 65 ; 4, 54, 67 ; cave, or vide, ne with the present subjunctive 
in 4, 3, 5 ; 4, 4, 6. Following is a complete list of the instances 
of ne with the perfect in Cicero's Letters, nearly all of which show 
great earnestness, either real or assumed : ad Att. 2, 5, 1 Etiam 
hercule est in non accipiendo non nulla gloria : qua re si quid 
eeo0an)y tecum forte contulerit ne omnino repudiaris ; ib. 5, 11, 
7 nam illam vofiavbpia (?) me excusationem ne acceperis; ib. 9, 9, 
1 Quod vereri videris ne mihi tua consilia displiceant, me vero 
nihil delectat aliud nisi consilium et litterae tuae ; qua re fac, ut 
ostendis : ne destiteris ad me quicquid tibi in mentem venerit 
scribere : mihi nihil potest esse gratius (Notice the emphatic 
position of words, indicative of strong feeling) ; ib. 10, 13, 1 
Epistola tua gratissima fuit meae Tulliae, et mehercule mihi; 
semper secum aliquam (?) adferunt tuae litterae. Scribes igitur 
ac, si quid ad spem poteris, ne demiseris. Tu Antoni leones 
pertimescas cave; ad Brut. 1, 16, 6 Me vero posthac ne commen- 
daveris Caesari tuo, ne te quidem ipsum, si me audies. Valde 
care aestimas tot annos, quot ista aetas recipit, si propter earn 
causam puero isti supplicaturus es ; ad fam. 7, 17, 2 Hunc tu 
virum nactus, si me aut sapere aliquid aut velle tua causa putas, 
ne dimiseris; ib. 7, 25, 2 Sed heus tu . . . secreto hoc audi, tecum 
habeto, ne Apellae quidem, liberto tuo, dixeris ; ad Quint. 1, 4, 5 
Sin te quoque inimici vexare coeperint, ne cessaris ; non enim 
gladiis tecum, sed litibus agetur; ad Att. 1, 9 ne dubitaris mittere 
(' Do not for a moment hesitate,' etc.) ; ib. 4, 15, 6 Veni in spec- 
taculum, primum magno et aequabili plausu — sed hoc ne curaris ; 
ego ineptus, qui scripserim ; ib. 7, 3, 2 Quin nunc ipsum non 


dubitabo rem tantam abicere, si id erit rectius; utrumque vero 
simul agi non potest, et de triumpho ambitiose et de re publica 
libere. Sed ne dubitaris quin, quod honestius, id mihi futurum 
sit antiquius ; ad Quintum fratrem 2, 10, 5 locum autem illius de 
sua egestate ne sis aspernatus (Cicero is here speaking of Caesar, 
which sufficiently accounts for his vigorous tone). In ad Att. 16, 
2, 5 Planco et Oppio scripsi equidem, quoniam rogaras, sed, si 
tibi videbitur, ne necesse habueris reddere, we should have 
expected the present. Here, however, it might be noticed that 
the first hand of the Medicean manuscript (M), the highest 
possible manuscript authority and in fact the only authority of 
much importance, omits the ne. In ad fam. 7, 18, 3 Tu, si inter- 
vallum longius erit mearum litterarum, ne sis admiratus, sis is 
probably the verb, admiratus being here used adjectively, as in 
ad Att. 9, 12, 2 sum admiratus ('I am surprised'), and in Off. 2, 
10, 35 ne quis sit admiratus etc. 

Most of the instances to be found, in the prose of classical 
times, of ne with the 2d person present subjunctive in prohibitions 
have been cited earlier in this paper. The following should be 
added to complete the list: Cic. Cato Maior 10, 33 ne requiras; 
ib. ad Att. 2, 24, 1 ne sis (perturbatus perhaps here used adjec- 
tively, like the following sollicitus and anxius). There are a 
large number of other passages that might well be explained 
as instances of the same use, e. g. ad Att. 14, 1, 2 Tu, quaeso, 
quicquid novi scribere ne pigrere (which Madvig, Opus. 2, 
p. 107, and Kiihner, Lat. Gram. II, §47, 8, actually explain as 
independent of quaeso); Phil. II 5, 10; pro Cluentio 2, 6 ne 
repugneiis etc. That ne with the present subjunctive is not 
more common in the best prose is due to an increasing fond- 
ness for the ^^'-construction. Ne with the present was a mild 
prohibition as compared with ne with the perfect, but it was 
less deferential and respectful than noli, and in dignified address 
noli accordingly became the regular usage. In early comedy 
there was comparatively little call for the more calm and dignified 
forms of expression, and there accordingly we find that noli is 
comparatively rare. It occurs in Plautus and Terence only in 
addressing some one who must be gently handled. It is found 
only where the tone is one of pleading — it never conveys an 
order, in the strict sense of that word. It is almost never used 
by a superior in addressing an inferior. In the two or three 
exceptions to this rule, the superior has some motive for adopting 


the mild tone. Those who wish to test the truth of these remarks 
are referred to the following complete list of the instances of noli 
in Plautus and Terence : Plaut. Am. 520 ; 540 ; Capt. 845 ; Miles 
372; 1129; Trin. 627; Asin. 417; Cure. 128; 197; 697; Most. 
800; Merc. 922; Poen. 367; 871; 1319; Persa 619; 831; True. 
664; Cas. II 2, 32; II 6, 35; Cist. I i, 59; I 1, 109; Ter. And. 
3 8 5J 68 5 5 Phorm. 556; Hec. 109; 316; 467; 654; Adelph. 781. 
As regards the different forms of prohibition in classical times, 
nothing can show more strikingly the difference in feeling between 
ne with the perfect subjunctive and noli with the infinitive than a 
comparison of the classes of verbs found in the two constructions. 
Of the 123 instances of noli in Cicero, 76 of them are used with 
verbs indicating some mental action, or some action which would 
be as unlikely to be accompanied by emotion on the part of the 
speaker, e. g. 'Do not suppose,' 'Do not be afraid,' etc. 1 In the 
Letters, 21 out of the 32 instances are verbs of this sort. Of the 
30 instances of cave with the subjunctive, 17 are of this sort. 2 In 
the Letters the proportion is 11 out of 18. A glance at the 
instances above cited of ne with the present subjunctive will show 
that most of the verbs in this construction also belong to the 
same class. We found the same state of things also in Plautus 
and Terence. Now, side by side with these facts put the fact 
that in the whole history of the Latin language, from the earliest 
times down to and including Livy, there are to be found in pro- 
hibitions expressed by ne with the perfect subjunctive only two, 
or at most three, verbs denoting mere mental activity, viz. ne 
dubitaris (Cic. ad Att. 7, 3, 2), ne metueritis (de rep. 1, 19, 32), ne 

1 Plane. 18, 44; 19, 46; 19, 47; 20, 50; 21, 51 ; 22, 52; 22, 53 ; Balb. 28, 64; 
Pis. 20, 46; 27,66; Marcel. 8,25; Ligar.11,33; 12,37; Phil. 2,28,69; 7,8, 
25 ; 12, 6, 14; de or. 2, 47, 194 ; 2, 61, 250; 2, 66, 268; Brut. 33, 125, 40, 148 ; 
nat. deor. 2, 18, 47; Cato 22, 79 ; Rose. Am. 24, 67 ; in Caec. div. 12, 39; Verr. 
2, 1, 16, 42; 2, 1, 49, 128 (twice); 2, 2, 11, 29; 2, 2, 51, 125; 2, 3, 5, 11 ; 2, 3, 
46, 109; 2, 4, 5, 10; 2, 4, 51,113 (twice); 2, 5, 5,10; 2, 5, 18, 45; 2, 5, 53.139'. 
de re pub. 1, 41, 65 ; 2, 3, 7; Orat. prid. quam in exsil. iret 1, 1 ; Tusc. disp. 
5, 5, 14; imp. Pomp. 23, 68; agr. 2, 6, 16; 2, 28, 77; Mur. 19, 38; 37, 80; 
Flacc. 20, 48 ; 42, 105 ; Sull. 16, 47 (twice) ; 27, 76 ; de dom. 57, 146 ; de 
harusp. responso 28, 62; ad Att. 1, 4, 3 ; 2, 1, 5 ; 5, 2, 3 ; 6, I, 3 ; 6, I, 8 ; 8, 12, 
!3; 9. 7, 5 I I 2 . 9! x 3. 29, 2 ; 15, 6, 2 ; 16, 15 ; ad Brut. 1, 13, 2 ; ad fam. 4, 4, 
3 ; 4. 5. 5 I 5, 21. 1 ; 7, 25. 1 5 12, 16, 1 ; 12, 33 ; ad Quint. I, 2, 4, 14 ; 3, 6, 7 

2 Ligar. 5, 14 ; 5, 16 (twice) ; de rep. I, 42, 65 ; de leg. 2, 3, 7 ; Tusc. disp. 
5, 7, 19; ad Att. 5, 21, 5 ; 7, 20, 1 ; 8, 15, A 2; 9, 9, 4; 9, 19, 1 ; 10, 13, 1 ; ad 
Brut. 1, 15, 1 (twice); ad fam. 7, 6 ; 7, 25, 2 ; 9, 24, 4. 


curaris (ad Att. 4, 15, 6). 1 The only other verbs (four or five in 
number) dealing with mental action distinctly involve also other 
sorts of action. These are ne sis aspernatus (ad Quint, fratrem 2, 
io, 5), ne asciveris neve fueris adsensus (Ac. 2, 40, 125), commotus 
ne sis (pro Mur. 31, 65), and ne repudiaris (ad Att. 2, 5, 1). There 
are not so many objections to regarding nee existimaveris in Livy 
21, 43, 11 as a prohibition as there would be in Ciceronian Latin, 
though it is extremely doubtful even here. In any case, nothing 
of the sort should cause surprise in Livy, as he marks the begin- 
ning of a general breaking up of the strict canons observed in the 
best period. Livy (3, 2, 9) even goes so far as to say ne timete, 
which, in prose, would have shocked the nerves of Cicero beyond 
expression. The almost entire avoidance, until after the Augustan 
period, of this whole class of verbs expressing mere mental 
activity in prohibitions expressed by ne with the perfect subjunc- 
tive, and its remarkable frequency in other forms of prohibitions, 
can, it seems to me, be explained only in one way. Verbs of this 
class are, from their very nature, such as would not often be 
accompanied with passionate feeling, and so are confined to the 
milder forms of expression. And this, it seems to me, goes far 
to establish my contention that ne with the perfect subjunctive is 
reserved for prohibitions that are prompted by uncontrollable 
emotion, or else that are intended to be as vigorous as possible 
in tone, either, as is generally the case, from some serious motive, 
or merely as a bit of familiar pleasantry. This tone is commonly 
one of commanding. Rarely it is one of earnest entreaty, though 
in such cases the prohibition is commonly introduced by noli. 
Noli with the infinitive is the expression best calculated to win 
the good-will of the hearer, as it merely appeals to him to 
exercise his own will (i. e. 'Be unwilling'), or to forbear using it ; 
while ne with the perfect subjunctive disregards altogether the 
will of the person addressed, and insists that the will of the 
speaker be obeyed. 

Cornell University. H. C. ELMER. 

^Ne necesse habueris reddere (ad Att. 16, 2, 5) is but poorly supported by 
manuscript evidence. Even if the reading is correct, as seems highly prob- 
able, the idea of reddere may be said to figure quite as prominently in the 
prohibition as that of habueris. Such expressions as ne vos quidem timueritis 
(Cic. Tusc. Disp. I, 41, 98), numquam putaveris (Sail. lug. no, 4) and nee 
putaveris (Cic. Acad. 2, 46, 141) represent very different uses, as I shall show- 
in Part II of my paper.